CN115731568A - Equipment safety management system combining wearable equipment and human body detection module - Google Patents

Abstract

The invention relates to an equipment safety management system combining wearable equipment and a human body detection module, which is characterized by comprising the following components: the staff confirming module is used for acquiring detection information aiming at the staff from the at least one detection module and confirming the staff in the operation object equipment according to the acquired at least one detection information; and a lock control signal output module that outputs a lock control signal to the work target apparatus based on at least one of a control signal input from the control module of the work target apparatus or a worker detection signal input from the worker confirmation module.

Description

Equipment safety management system combining wearable equipment and human body detection module

Technical Field

The present invention relates to a device security management system, in particular to a security management system incorporating a wearable device and a human detection module.

Background

For the general manufacturing industry, it is the industrial characteristics that medium and large-sized equipment is arranged, and under the working environment of the large-sized equipment in operation, the situation that the equipment is injured or personal accidents occur because safety rules are not obeyed in the internal working process of the equipment often occurs.

In particular, the equipment worker needs to enter the equipment to perform work due to reasons such as process monitoring, maintenance for damage, and the like, material input, and periodic inspection, and in principle, the equipment must be driven after the equipment is stopped.

However, since there is a possibility that an accident may occur, it is necessary to pay attention to the fact that the work is performed without stopping the equipment or that another worker who stops the equipment but does not know the fact restarts the equipment.

In order to prevent such a safety accident, a change is made in the apparatus in which the apparatus cannot be automatically restarted at all by separately providing an apparatus restart key, and a function of detecting an open state of an access door and automatically stopping driving is carried out, and even so, the safety accident is continuously generated, and thus an effective solution is still required.

Disclosure of Invention

Technical problem

In order to solve the problems described above, the following provides a device security management system combining a wearable device and a human body detection module.

The technical problems to be solved by the present invention are not limited to the above technical problems, and other technical problems not mentioned can be clearly understood by those skilled in the art to which the present invention pertains from the following descriptions.

Technical scheme

A device security management system combining a wearable device and a human body detection module according to an embodiment of the present invention for solving the above-described problems is characterized by comprising: the staff confirming module is used for acquiring detection information aiming at the staff from the at least one detection module and confirming the staff in the operation object equipment according to the acquired at least one detection information; and a lock control signal output module that outputs a lock control signal to the work target apparatus based on at least one of a control signal input from the control module of the work target apparatus or a worker detection signal input from the worker confirmation module.

In addition, the present invention is characterized in that the lock control signal output module outputs a lock control signal for stopping the operation of the work target apparatus when the lock signal of the work target apparatus in the control signal is received or the operator is confirmed by the operator confirmation module, and the lock control signal output module outputs a lock control signal for starting the operation of the work target apparatus when an unlock (unlock) signal of the work target apparatus in the control signal is received and it is confirmed that the operator is not present in the work target apparatus.

In addition, the present invention is characterized in that the at least one detection module includes a wearable device (wearable device), the wearable device is activated by recognizing a wearing state of the worker, and when the wearable device receives a lock code associated with the work target device from the worker, the worker confirmation module determines that the worker is confirmed not to be in the work target device when the unique identifier of the wearable device and an unlock code corresponding to the lock code are recognized. In addition, the at least one detection module may include a wearable device measurement module including three or more antennas, and the wearable device measurement module may compare the position of the worker with the area of the work target device based on triangulation based on a distance from each of the three or more antennas to the wearable device.

In the present invention, the at least one detection module includes an image-based detection module that determines that the worker is present if an extracted image having a similarity of a predetermined level or higher with respect to an image database learned based on a data set (data set) composed of a predetermined plurality of individual images is present in a first image region on a predetermined number or more of consecutive frames.

In addition, the at least one detection module may include an image-based detection module, and the image-based detection module may determine that the worker is present when an image different from a predetermined image is detected in the second image region.

ADVANTAGEOUS EFFECTS OF INVENTION

According to the embodiment of the invention, the safety management can be effectively carried out through the equipment safety management system combining the wearable equipment and the human body detection module.

The effects obtainable by the present invention are not limited to the above effects, and other effects not mentioned can be clearly understood by those skilled in the art to which the present invention pertains from the following description.

Drawings

The accompanying drawings are included to provide a part of the detailed description of the invention and explain the technical idea of the invention together with the detailed description, and to provide an embodiment of the invention as a part of the detailed description.

Fig. 1 is a structural diagram of a device security management system according to an embodiment of the present invention.

Fig. 2 is a reference diagram for explaining signal processing aspects of the device security management system according to an embodiment of the present invention.

Fig. 3 is a reference diagram for explaining the operation steps of the device security management system using a wearable device according to an embodiment of the present invention.

Fig. 4 is a reference diagram for explaining the operation in the image detection module according to an embodiment of the present invention.

Detailed Description

The present invention is not limited to the description of the embodiments described below, and it is apparent that various modifications can be made without departing from the technical spirit of the present invention. In the description of the embodiments, technical contents that are known in the technical field to which the present invention pertains and that do not directly relate to the technical gist of the present invention will be omitted from the detailed description.

On the other hand, the same structural elements in the drawings are denoted by the same reference numerals.

In the drawings, some components may be enlarged, omitted, or simply illustrated. This is for clearly explaining the gist of the present invention by omitting unnecessary descriptions that are not related to the gist of the present invention.

Fig. 1 is a structural diagram of a device security management system according to an embodiment of the present invention.

The equipment safety management system of the present invention may include a worker confirmation control device 110 to which the present invention is applied and a work object equipment 130 to which the worker confirmation control device is coupled/inserted.

The operator confirmation control apparatus 110 of the present invention may be formed with an operator confirmation module 111 and a lock control signal output module 113.

The worker confirmation module 111 can obtain detection information on a worker within the working radius of the working object apparatus from various modules including: the wearable device and the wearable device identification module can receive confirmation of the operation content by identifying the wearing/non-wearing state of the operator or displaying the operation content to the operator; the image shooting module is composed of a camera, a monitoring Closed Circuit Television (CCTV) and the like; an image-based detection module that may detect whether a worker is present by receiving an image from the image photographing module; and a detection sensor module operating with infrared rays or the like. That is, in the case where the detection information is obtained from at least one detection module, the worker confirmation module may confirm or make a judgment as to whether or not a worker is present in the work target apparatus.

Furthermore, the detection module connected to the staff confirmation module may be preferentially activated, e.g. the wearable device identification module will be activated in case the wearable device is activated, or the wearable device may be deactivated in case the image based detection module is activated. Conversely, in the case where an error or the like occurs in worker recognition by the image-based detection module or the like, the wearable device or the human body detection sensor may be activated. That is, even if the worker confirmation module uses a plurality of detection modules, it is possible to activate or deactivate part of the detection modules by considering the state of the sensor, the state of the worker, the battery consumption, and the like.

The lock control signal output module 113 may output a lock signal/unlock signal to a lock target device 133 among the work target devices based on a control signal input from an existing control module 131 that controls the work target device 130 or an operator detection signal input from the operator confirmation module 111.

In particular, the present invention is characterized in that the lock control signal output module controls only the lock signal among the received control signals by intermediately receiving a control signal transmitted from the control module 131 provided in the work target apparatus or existing outside the work target apparatus (for example, a central control server or the like) to the lock target apparatus 133 provided in the work target apparatus. During the work, a worker may transmit a test signal to the work target equipment to confirm whether or not the work target equipment is normally maintained. Therefore, blocking all signals transmitted from the control module in the middle may cause problems of lowering the inspection efficiency of the work object apparatus of the worker, work delay, and the like.

Therefore, the lock control signal output module 113 of the present invention can determine whether to selectively transmit the control signal to the work target apparatus or to output the lock control signal related to locking or unlocking to the work target apparatus after confirming whether the control signal received from the control module is the lock signal or the unlock signal of the work target apparatus.

In addition, the lock control signal output module 113 of the present invention determines not only the control signal input from the control module for the work target apparatus but also the worker detection signal of the worker confirmation module 111 to confirm whether the worker is confirmed and outputs the lock control signal. In this regard, the lock control signal output module 113 can separately determine the condition of the lock target apparatus and the condition of the unlock target apparatus.

The lock control signal output module 113 may stop the operation of the work target apparatus whenever any one of the lock signal condition of the work target apparatus in the reception control signal or the condition of the worker being confirmed by the worker confirmation module occurs. This is to prevent an accident from occurring to the worker by preventing the work target apparatus from working erroneously.

However, it is preferable that, in the case of unlocking the work target apparatus, the work target apparatus should be started after all the detection signals and the control signals are confirmed. This is to prevent an accident that may occur when an unlock signal is input when it is not possible to externally confirm whether or not a worker is present in the work target apparatus.

Therefore, the lock control signal output module may output the lock control signal for starting the operation of the work target apparatus when the unlock signal of the work target apparatus is received from the control signal received from the control module and the worker confirmation module confirms that the worker is not present (i.e., the worker is absent) in the work target apparatus.

In addition to the operation of outputting the lock control signal, the lock control signal output module 113 according to the present invention may perform an operation of causing a warning lamp provided outside the apparatus to give an alarm by using a warning module while causing the apparatus to be stopped in an emergency, or may share an emergency between workers by using a wearable apparatus, in a case where the apparatus to be operated is in an abnormal state (for example, although the entry/exit door is always in a closed state, which is a normal state of the apparatus to be operated, but is in an open state). That is, the reminder module may display a reminder consisting of a haptic effect, an auditory effect, a visual effect to the worker with a specific code that may enable the worker to identify the locked state of the manufacturing device. For example, the reminder module may notify that a worker is present in the work object apparatus or the work object apparatus is in an abnormal state through at least one of a light, a lamp, a display, a speaker, and a wearable apparatus associated with the manufacturing apparatus.

In the present invention, the detection module, which can transmit the worker detection signal by being connected to the worker confirmation module, may be formed in various forms. For convenience of description, the wearable device measurement module, the image-based detection module, and the human body detection sensor are mainly used as the center, but additional devices such as an internet of things (IoT) sensor that can transmit a detection signal indicating the presence of a worker may be added to the worker confirmation module may be added to the present invention. In the present invention, the wearable device 121 may transmit the worker detection signal to the worker confirmation module 111 by being connected to the worker confirmation control apparatus 110.

For example, in the case where the worker is identified by a wearable device (e.g., a smart band, a work helmet, a wireless headset, etc.) having a unique identifier, the wearable device may be utilized to identify whether the worker is working on the lock object device.

In this case, the wearable devices 121 may be configured to have respective unique identifiers, and a single unique identifier may be configured to match a particular worker. Therefore, it is possible to display information on the contents of a previously planned work, a work target device, and the like to a specific worker, or to grasp information on whether or not the specific worker performs a work activity, a work time, the work target device, and the like, by a manager, and it is possible to grasp an abnormal situation in which the worker is close to the work target device, which is not related to the work target device.

The wearable device 121 of the present invention can be activated by recognizing the wearing state of a specific worker, and in the case where the worker is recognized by the activated wearable device, the lock of the work object device is released only by confirming the unique identifier of the activated wearable device.

This is because the case where the work target apparatus is locked by the specific worker is definitely determined by the worker confirmation module as the work being performed on the target apparatus. For example, when the wearable device receives a lock code related to the work object device from the worker, the worker confirmation module may determine that the worker is not within the work object device if the unique identifier identifying the wearable device and the unlock code corresponding to the lock code are confirmed. However, the lock code may be a code related to a specific work target apparatus, a code based on a specific maintenance plan, or a code related to a staff group to which a specific staff member belongs, or a lock code/unlock code may be shared by a plurality of staff members.

For example, when the worker wears the wearable device 121 and activates the wearable device, if the start of the maintenance work on the work target device is notified in such a procedure, the worker confirmation apparatus stores the unique identifier of the wearable device notifying the start of the maintenance work, and unlocks the work target device only when the maintenance work on the work target device is ended by the wearable device matching the unique identifier. The worker can conveniently stop the device before starting the operation by using the wearable device, and can prevent safety accidents caused by accidental driving of external workers by giving a re-drivable right only to the worker who implements the stopping driving.

In the present invention, the lock/unlock of the lock target device 133 is realized only by a specific wearable device, and the operator wearing the wearable device may be periodically notified to prevent the operator who starts the work from unnecessarily stopping the lock target device because the operator is not unlocked. For example, the worker is periodically notified of the locked state of the lock target apparatus by vibration or a warning sound of the wearable apparatus, so that the worker can be reassured and the work target apparatus can be prevented from being in a stopped state by carelessness of the worker even when the work on the work target apparatus is ended. Further, when the worker notifies the worker of the start of the work to the work target apparatus through the wearable apparatus to confirm the control device or the like, the start of the work may be transmitted to a colleague related to the maintenance of the work target apparatus. If a human detection sensor/image capturing module (camera) or the like related to the lock object device is activated, the wearable device may not be activated. As described above, this is because, in the case where one detection module normally identifies a worker, it is not necessary to identify the worker repeatedly by the other detection modules, and it is necessary to reduce battery consumption of the wearable device. However, even in this case, it is preferable that, if the worker confirmation module recognizes the start of the job by the wearable device, the lock target device is unlocked only when the wearable device is activated and the job is ended.

Further, in the present invention, in the case where it is set that the wearable device 121 is connected to the worker confirmation control apparatus 110, in order to grasp the exact position of the worker, the wearable device measurement module 123 may be additionally activated.

The wearable device measurement module 123 of the present invention can measure the position of the worker through at least one antenna. That is, the distance between the antenna and the wearable device can be measured by a single antenna constituting the wearable device measurement module, and the position of the wearable device can be confirmed by triangulation using a plurality of antennas, whereby the presence or absence of a worker inside the lock target device can be confirmed. A detailed description of this will be described in more detail in the following description in relation to fig. 3.

In the present invention, as one of the detection modules connected to the worker confirmation module, the image-based detection module 125 may be utilized. The image-based detection module 125 analyzes the image extracted in the specific image area, and may determine that a worker is present in the case of similarity to the predetermined image, and may determine that an abnormal situation occurs and the lock target apparatus stops operating if a different image other than the predetermined image is detected in the specific image area. A detailed description of this will be described in more detail in the following description in relation to fig. 4.

Further, in the present invention, the entry of the worker can be confirmed by the human body detection sensor module provided inside/outside the lock target apparatus. Such a human body detection sensor may utilize various sensors such as an infrared sensor, an ultrasonic sensor, an ultraviolet sensor, a motion recognition sensor, etc., and may transmit a worker detection signal to a worker confirmation module by constituting an internet of things (IoT) -based network. In addition, in the present invention, since the worker is identified by the plurality of detection modules, a priority order may be set for the efficiency of the work sensor, and for example, the human body detection sensor may be set to be activated only when the wearable device is not activated and the image processing by the image detection module becomes difficult. Therefore, when a specific detection module is in an abnormal state, an attempt is made to activate other detection modules connected to the worker confirmation module, and when all detection modules connected to the worker confirmation module are not activated, an unlock signal is transmitted from the control module of the work object apparatus, and even when the worker confirmation module makes a judgment that the worker is not confirmed, a lock control signal for stopping the work object apparatus is output to the work object apparatus. This is to prevent the worker from confirming that the module made an incorrect judgment due to an error of the detection module.

Therefore, according to the present invention, while minimizing the stop time (i.e., the stop operation time) of the equipment to be operated, it is possible to provide the worker with a safe environment that ensures that the equipment is not driven until the necessary operation is finished, and the manager or the co-worker can share the current driving state and the operation history information of the equipment in real time and can achieve an effect of easily grasping the situation of the site.

Fig. 2 is a reference diagram for explaining signal processing of the wearable device and the device security management system according to an embodiment of the present invention from the aspect of signal processing. In the description related to fig. 2, for convenience of description centering on signal processing, description will be given by using the name of a transceiver.

In the case where the worker-use transceiver 210 (e.g., wearable device) is activated by an action of being worn by a worker or the like, the worker-use transceiver may transmit a signal related to device entry/release to a control module, a worker confirmation control apparatus, or the like, and may receive event information related to the lock-target device (e.g., start/end of work by other workers, state of the lock-target device). However, for example, with a wearable device, in the case of entering the inside of a work object device, there may occur a case where a signal becomes poor or the signal strength is relatively poor due to a problem of a battery or the like, and there may occur a problem that it is difficult to receive a signal or event information related to entry/release of the device.

Accordingly, the device security management system of the present invention may further include a repeater 220. The repeater 220 may amplify and transmit signals received from the staff-use transceiver to the equipment-use transceiver or may transmit amplified signals from the equipment-use transceiver to the staff-use transceiver. That is, each repeater is used for a signal transceiver for a device used for a wearable device, a sensor constituting an internet of things (IoT) network, and a work target device, and each repeater amplifies a signal transmitted to the repeater using a predetermined protocol, thereby increasing reception sensitivity and a reception distance, and thus enabling transmission and reception of a signal with a transceiver for a worker, a work target device, or another repeater.

The device-use transceiver 230 may transmit to the lock control signal output module by converting a signal received from the repeater 220 or may transmit to the repeater/staff-use transceiver by converting a signal received from the lock control signal output module. The device-wide transceiver may be included in the staff member confirmation module, and the staff-worn transceiver (e.g., wearable device) may utilize a variety of wireless communication protocols, according to which it may be necessary to translate into a signal that may be recognized by the locking control signal output module. Therefore, in the present invention, as the transceiver for the device is used, the wearable device may use various protocols such as Long Term Evolution (LTE), wireless fidelity (WIFI), bluetooth Low Energy (BLE), low power local area network wireless standard (LORA), and the communication method is not particularly limited. Further, signal transmission from the device using a plurality of communication protocols to the lock control signal output module 240 can be stably performed only by the worker performing hardware/software update with the transceiver.

The lock control signal output module 240 may transmit a driving/stopping instruction to the work object apparatus based on at least one of signals transmitted through the control module and the worker-use transceiver, and may transmit to the worker-use transceiver in the case where an event occurs in the work object apparatus. Further, information on entry/unlocking of the work target device by the worker, a state of the work target device (an activated state, door (Door) open, failure, etc.) is transmitted to the control module based on the signal recognized or transmitted by the device transceiver 230, thereby helping the manager to grasp the information on the work target device.

In this case, the lock control signal output module 240 of the present invention transmits a signal to the work target apparatus by using Power Line Communication (PLC), and since the manufacturing apparatus may have a size larger than a signal radius of wireless communication, and the signal strength may be deteriorated or may be disturbed according to a mechanical structure of the manufacturing apparatus, etc., the work target apparatus is controlled by using the power line communication, and thus the control command may be stably transmitted to the work target apparatus.

The control module 260 is a worker confirmation control device including a lock control signal output module, and can receive information such as the current driving/stopping status and the operation history of the work target equipment, store the information in a database or the like, display the information to a manager, and provide a manager function such as forced re-driving or data initialization of the work target equipment.

Fig. 3 is a reference diagram for explaining the working steps of the device security management system using the wearable device and the wearable device measurement module according to an embodiment of the present invention.

According to an embodiment of the present invention, the staff member may input the locking code through the activated wearable device (step S310). When a worker enters the work target device and the work start situation of the worker is indicated by the wearable device, the corresponding work content can be displayed on the wearable devices of other workers sharing the same lock code, and if the work target device is in a driving state, the lock state can be changed to a lock state in which the work target device is stopped or cannot be restarted. Further, a plurality of workers who share a lock code relating to the work object apparatus in the locked state can periodically receive a signal relating to the state of the work object apparatus.

In this regard, the device security management system may determine whether additional activation of the wearable device measurement module is required if the wearable device is activated (step S320), and may determine the location of the worker by using the wearable device measurement module if the wearable device measurement module is activated. The wearable device measurement module may include at least one antenna, and the form of the measured information in the wearable device measurement module may be changed according to the number of antennas identifying the wearable device. For example, in a wearable device measurement module, with one antenna, only a wearable device can be identified, and with two antennas, candidate locations for the wearable device can be calculated by considering the distances that the respective antennas identify to the wearable device. In the case where three or more antennas are used, the position of the worker can be measured by triangulation or the like based on the distance to the individual antenna. In particular, in a manner of identifying a location by triangulating a distance of the wearable device through a single antenna, a communication protocol using Radio Frequency (Radio Frequency), bluetooth, wireless fidelity (WIFI), or the like may be applied thereto.

Hereinafter, the description will be mainly given of a case where the position of the worker is measured by using three or more antennas. In the case where the wearable device measurement module of the device security management system is activated, the position of the worker is determined using three or more antennas (step S325).

For example, the position of the worker identified from the wearable device and the area of the work target device may be compared based on triangulation based on distances from three or more antennas to the wearable device, respectively. In this case, the single antenna may be provided at a position predetermined by a manager, but may also include a Global Positioning System (GPS) sensor in such a manner that the position can be automatically recognized in combination with the work object apparatus.

The worker confirmation module of the security management system may determine whether a worker is present in the work object apparatus (step S330). Even if the operator does not indicate his/her work start situation by the wearable device when he/she enters, the operator can be confirmed whether he/she enters by the image-based detection module or the human body detection module combined with the wearable device. In this case, when all or a part of a human body is detected from the work target apparatus or when the worker cannot be normally detected by the worker detection module (e.g., a camera, a monitoring cctv, or a human body detection sensor) (e.g., when the worker is blocked by the camera), it is determined that the worker has entered, and then the operation of stopping the work target apparatus is executed.

When the position of the wearable device is measured, the control server or the like may map the position of the wearable device on the layout of the work target device to determine whether or not the work target device is located inside the lock target device and the current position of the worker.

In this case, the worker confirms that the control device receives a signal from the control module related to the work target apparatus in addition to the worker detection module (step S340).

In contrast, a lock control signal for starting or stopping the operation of the lock target device may be output based on the lock signal among the control signals received from the control module or the operator detection signal received from the operator confirmation module (step S350).

When the lock control signal is used to lock, the device security management system stops the lock target device to be locked among the work target devices (step S360).

When the worker ends the job for the job target device (step S370), the job target device may be driven. For example, when there is an external drive attempt, if it is determined that it is safe normally after human body detection inside the device and that there is no job notification by the wearable device, the device may be started. Also, if the corresponding worker does not release the job through the wearable device, the device in the locked state cannot be restarted, and therefore, if the unlock code is not confirmed through the wearable device, the worker is regularly notified that the device is currently in the locked state. If the wearable device fails to indicate the completion of the work when the operator fails to take his/her place, the wearable device cannot be restarted because the state of the wearable device is not changed from the locked state.

In this regard, the device security management system will determine the unique identifier of the activated wearable device and the unlock code corresponding to the lock code (step S380). That is, when the worker leaves the work area, if the wearable device indicates that the worker has finished his work, the wearable device of the other worker sharing the same code displays the corresponding work content, and the worker confirmation module of the work target device confirms that the worker is absent.

If the detection cannot be normally executed by the staff member confirming module, the state of the equipment to be locked is judged to be in an abnormal state by assuming the existence of staff members, and the locked state is maintained.

When it is determined that the worker is not present and the unlock signal is input from the control module, the lock control signal output module outputs a lock control signal for unlocking the lock control module to unlock the work target apparatus (step S390). That is, when the worker tries to drive the vehicle from the outside after leaving the field, if the completion of the work is confirmed by the wearable device or the like, the completion of the work is indicated to the other worker, and the lock target device is restarted after the in-device detection module confirms the normal condition.

Fig. 4 is a reference diagram for explaining the operation in the image detection module according to an embodiment of the present invention.

The image-based detection module of the present invention may stop the operation of the job target apparatus in a case where it is detected that a single image stored in advance exists on a plurality of consecutive frames within the image area or an image for detecting a change to a predetermined image is detected. Among them, a single image stored in advance may be learned by employing artificial intelligence, for example, a detection engine for an image-based detection module may be learned using a Convolutional Neural Network (CNN) as a deep learning algorithm.

For example, the data set to be the human body recognition learning data in the present invention includes not only the whole body image of a person but also data of a part of the body mainly used by a worker, so that it can be learned that detection can be realized as long as the worker is reaching for work or only a part of the body enters the apparatus. In particular, the image classification method applied to the present invention is configured such that, when a driving unit (motor, robot, etc.) of a work target apparatus is recognized in an image, it is determined whether or not the extracted image belongs to a single image (e.g., human) stored in advance, instead of tracking an object that generates a dynamic motion, a high recognition rate for a worker is exhibited, and the recognition rate can be continuously increased by learning false detection/non-detection condition data using a convolutional neural network. Therefore, the detection engine which performs learning through deep learning can interpret images transmitted by the camera in real time to judge whether a person is inside the equipment to be operated, and even if only a part of the body of a worker appears, the person can be identified by the image detection module according to the visual angle/shooting range of the camera. That is, the detection engine of the present invention can classify various body parts by learning various body parts such as hands, feet, and head of a person and improve the accuracy of determining that a person enters.

Therefore, in the present invention, the detection engine that performs learning by artificial intelligence can be used to perform real-time image recognition on the captured image, and when a body is recognized in a predetermined number or more of consecutive frames, it can be determined that a person has entered the interior. The number of frames to be used as a basis for judgment by the detection worker and the level of similarity for judging the extracted image may be set by the administrator in advance, but may be gradually increased along with the learning data of the detection engine.

That is, in the case of judging the movement (entrance ) of the worker by the camera, there is a problem that the possibility of undetected (classifying entrance as not-entered) and false detection (classifying non-entered as entered) increases due to structural reasons, and the image-based detection module of the present invention can immediately perform emergency stop judgment upon detection of a body and can set a specific area in an image photographed by the camera or the monitoring cctv as an area where the worker is not allowed.

Part (a) of fig. 4 is a reference diagram for explaining a scheme of confirming whether or not there is a worker in an image area set in such a manner that the degree of similarity is judged by the extracted image.

In part (a) of fig. 4, if there is an extracted image having a similarity to an image database learned based on a data set (data set) composed of a predetermined plurality of individual images, which is equal to or higher than a predetermined level, in a specific image region of a predetermined number or more of consecutive frames, it can be determined that a worker is present.

The image capture module (e.g., a camera, a monitoring closed circuit television, etc.) transmits to the image-based detection module by continuously capturing, in real time, an area within the work object apparatus where the worker is likely to be present. In this regard, the image-based detection module may extract a single frame from the picture transmitted by the image photographing module to cluster (cluster) data for image classification. For example, a single frame may be analyzed in an image-based detection module, and the analyzed result values may be output in the form of an image and data, respectively. However, a frame related to one image corresponds to less than 0.1 second in a time unit, and therefore, a scheme of stopping the apparatus based on a result value of a single frame is not preferable, and it is preferable that it should be converted into accumulated data to determine whether or not a person actually enters the apparatus.

Therefore, taking the section (a) of fig. 4 as an example, when 15 frames Per Second (fps, frame Per Second) of video is analyzed, the results (images, data) are presented as the result values of "person 1 is 92%, person 2 is 72%, and table is 100%", and in a single Frame, "person 1 and person 2" are two individuals classified as bodies, and therefore, when the threshold (threshold) set by the administrator is reached or more, the image-based detection module can determine that a person is detected. For example, in the case where the threshold is set to 75%, in a single frame, the individual "person 1" can be judged as a body, and the remaining individual "person 2" will not exhibit a similarity degree equal to or higher than the threshold, and thus can be analyzed using an additional frame. In this case, the similarity of individuals can be judged with a high threshold in a single frame and with a relatively low threshold in successive frames, and whether or not there is a person can be judged more efficiently in this manner.

That is, in the case where it is determined that the data is insufficient or it is determined that the data is not a human body, the apparatus safety management system can determine whether or not the work target apparatus can be operated by determining an additional frame. Further, in the case where the image-based detection module detects the entrance of a person by accumulating data, an emergency stop instruction may be transmitted to the lock target apparatus. Further, the image in which the stop record of the work target device and the determination that someone has entered the image based on the performance of the image detection module can be separately stored.

Fig. 4 (b) is a reference diagram for explaining a case where a specific region of an image is set as a region where a worker is not allowed to enter. If an image different from the predetermined image is detected in the specific image, it can be determined that the worker is present. For example, when an image different from the image of the device is detected only in the image area of the device, it is possible to prevent an accident caused by an accidental entry of a worker by temporarily stopping the work. That is, performing worker detection using artificial intelligence for all areas in an image may cause a load on the calculation of the control server, and in order to perform detection more efficiently, the job target apparatus may be stopped as long as there is a different image in a specific image area.

And, the image-based detection module may perform similarity determination with respect to an image extracted from a specific region in one image through the detection engine while setting to detect different images with respect to the remaining regions, so that it is possible to effectively prevent an accident of a worker.

The embodiments of the present invention illustrated in the drawings and described in the specification are only specific examples disclosed for simply explaining the technical contents of the present invention and helping to understand the present invention, and do not limit the scope of the present invention. In addition to the embodiments disclosed, it is obvious to those skilled in the art that other modifications based on the technical idea of the present invention can be implemented.

Industrial applicability

The device security management system combining the wearable device and the human body detection module as described above can be applied to various industrial fields.

Claims (6)

1. An equipment safety management system combines wearable equipment and human body detection module, its characterized in that includes:

a worker confirmation module that acquires detection information for a worker from the at least one detection module and confirms the worker in the operation target device based on the acquired at least one detection information; and

and a lock control signal output module which outputs a lock control signal to the work target apparatus based on at least one of a control signal input from the control module of the work target apparatus or a worker detection signal input from the worker confirmation module.

2. The device security management system according to claim 1,

the lock control signal output module outputs a lock control signal for stopping the operation of the work target apparatus when the lock signal of the work target apparatus in the control signal is received or when the worker is confirmed by the worker confirmation module,

the lock control signal output module outputs a lock control signal for starting operation of the work target apparatus when the unlock signal of the work target apparatus in the control signal is received and it is confirmed that the worker is not present in the work target apparatus.

3. The device security management system according to claim 1,

the at least one detection module comprises a wearable device,

activating the wearable device by identifying the wearing state of the worker,

when the wearable device receives a lock code associated with the operation target device from the worker, the worker confirmation module determines that the worker is not in the operation target device if the unique identifier of the wearable device and the unlock code corresponding to the lock code are recognized.

4. The device security management system according to claim 3,

the at least one detection module comprises a wearable device measurement module consisting of more than three antennas,

the wearable device measurement module compares the position of the worker with the area of the work target device based on triangulation based on distances from each of the three or more antennas to the wearable device.

5. The device security management system according to claim 1,

the at least one detection module comprises an image-based detection module,

the image-based detection module determines that the worker is present if an extracted image having a similarity to an image database learned based on a data set including a predetermined plurality of individual images is present at or above a predetermined level in a first image region in a plurality of consecutive frames of a predetermined number or more.

6. The device security management system according to claim 1,

the at least one detection module comprises an image-based detection module,

if an image different from the predetermined image is detected in the second image area, the image-based detection module determines that the worker is present.

Images